PROJECT SUMMARY The availability of antiretroviral therapy (ART) has markedly reduced leading causes of HIV-related mortality and morbidity in Africa, including tuberculosis (TB). Yet TB remains the leading cause of death among people with HIV. Use of TB preventive therapy (TPT) reduces TB incidence and death, even among people receiving ART. Although the benefits of TPT among people with HIV have been known for 30 years and international and national guidelines provide clear prescribing recommendations, TPT is poorly prescribed. Overall in low and middle income countries 10-30% of people with HIV eligible for TPT receive it. For those who do receive TPT along with ART, adherence is generally good. Multiple system and provider level barriers appear to be driving anemic TPT prescribing. Several of these barriers have caused TPT prescribing to be the exception rather that the routine ? barriers based on complex procedures implemented for inaccurate concerns. Primary among are efforts to completely ?rule-out? active TB and concerns for potential liver problems with TPT. Notably, a study focused on more complex (and more sensitive) TB diagnosis resulted in increased mortality in the arm with greater TB diagnosis. That arm also had delayed and lower TPT prescribing. This study seeks to use choice architecture to make TPT prescribing the usual or ?default? with not-prescribing occurring only when the clinician has a real concern (e.g. high concern for TB disease). The effectiveness of the choice architecture-based implementation strategy will be compared to the usual implementation in a cluster- randomized trial. Clinics will be the unit of randomization with all patients receiving services at a study clinic receiving uniform TPT implementation. The primary outcome will be the proportion of patients initiating ART who also receive TPT. The underlying concept of choice architecture is that optimizing decision making can lead to reduced cognitive load. Thus we propose to compare cognitive load regarding TPT prescribing between study arms. We will also assess congruence of the prescribing approach with clinic work flow, acceptability to providers, implementation measures, and patient-level implementation (e.g. receipt of TPT, adherence, patient reported problems). This study has the potential to lead the way in reshaping the delivery of TPT and other routine services in clinics in South Africa and similar settings. Should this implementation strategy prove effective it will contribute to national and global goals to reduce HIV-associated mortality and TB incidence.